Data Interoperability: Lessons from UN Statisticians
21 February 2020
By Francesca Perucci, Chief of the Development Data and Outreach Branch and Luis Gonzalez Morales, Chief of the Web Development and Data Visualization Unit, Statistics Division, UN Department of Economic and Social Affairs
The 2030 Agenda for Sustainable Development puts data and statistics at the center of national and global action. But to unleash the full potential of data to achieve the Sustainable Development Goals (SDGs), all stakeholders must be more collaborative and integrated. This is because no single entity can control all the data that is required to implement and monitor progress towards achieving the Global Goals and leaving no one behind.
Increased collaboration and integration are only possible when decentralized data systems “work together” and are part of a network of a seamless data supply chain that includes many different producers and users of data at the national, regional and global levels. This is where interoperability - a process that allows data-driven organizations to share with each other the right data, in the right format, at the right time – is so important. Interoperability standards and best practices allow data to be presented in formats and structures that are easily comparable and integrated so that client systems can access data more readily. In this way, interoperability can help internal data processes become more efficient and streamlined, in turn playing a key role in facilitating data sharing for decision making.
The Statistics Division of the UN Department of Economic and Social Affairs (DESA), in close collaboration with many other members of the UN family, has for decades helped build statistical capacity in Member States by compiling and disseminating internationally comparable official statistics and coordinating a global statistical system. All these efforts have been based on standards and methodological guidelines for official statistics agreed by the UN Statistical Commission, the main intergovernmental body in statistics. In all these areas of our work, data interoperability has always been central to the development and maintenance of international statistical classifications, the provision of technical assistance for the implementation of data and metadata exchange standards, and the use of cutting-edge technologies to bring official statistics to the semantic web.
Today, interoperability is more of an organizational than a technological challenge. The speed at which the organization and governance of data systems and workflows evolve is still too slow compared to the speed of change in technology, and we are often in a catch-up mode as data managers. Many organizations still run multiple data platforms and systems that are poorly integrated with one another. But the reason for this fragmentation is not necessarily lack of knowledge or skills to implement technical standards for data interoperability, but rather lack of time and resources to focus on data innovation. It reflects how difficult it is to change course in the way day-to-day operations are carried out when staff have to deal with a continuous demand for new data products while keeping key legacy systems running.
Moreover, each department or programme s a complex organization in itself, with its own IT applications, technology, culture, data definitions, priorities, and, in some cases, different mandates and oversights. As a result, data items as simple as country codes are often defined (or re-defined), without consideration of interoperability issues down the road.
To enhance data interoperability in our organizations, we require a common set of data structure definitions, standard data models and API specifications, as well as a new generation of front-end applications that focus on facilitating sharing and integration of data using semantic-web infrastructure and similar technologies. But above all, we need sound data governance and management practices. And we must think beyond data availability and data access – which are huge challenges in their own right—to also focus on enhancing re-usability and impact of data in different contexts.
Fortunately, we are gradually seeing more concerted efforts towards building a truly global data architecture, where different data-driven organizations strive to speak a common “data language” based on simple but clear interoperability principles and standards, while considering the needs of local communities across stakeholder groups. Our challenge is to make every statistical output—from national statistical offices or from international organizations—more findable, accessible, interoperable and easily usable, across organizational and geographic boundaries as well as across disciplines and areas of expertise, with the ultimate goal to improve people’s lives.
Part of these efforts include the work of the Collaborative on data interoperability for sustainable development, co-convened by DESA’s Statistics Division and the Global Partnership for Sustainable Development Data. The collaborative launched Interoperability: A practitioner’s guide to joining-up data in the development sector, which brings together best practices from across the development sector highlighting the value that interoperable data brings to decision-making. The first version of the Guide was launched at the UN World Data Forum in Dubai in October 2018 and was endorsed by the United Nations Statistical Commission at its 50th Session in March 2019.
As we move forward in 2020, there are many signs that point towards wider adoption of standards and best practices for data interoperability both at the UN and beyond. These trends are promising given the importance of interoperability in facilitating data sharing for improved decision making.
*The views expressed herein are those of the author(s) and do not necessarily reflect the views of the United Nations.